test(ui): end-to-end AppTest tests for all 9 tabs; coverage floor in CI

.github/workflows/ci.yml

@@ -44,8 +44,7 @@ jobs:
 
       - name: Run tests with coverage
         run: |
-          # TODO: add --cov-fail-under=70 once coverage reaches target
-          pytest tests/ -v --cov=catlab --cov=app_ods --cov-report=term-missing --cov-report=xml
+          pytest tests/ -v --cov=catlab --cov=app_ods --cov-fail-under=81 --cov-report=term-missing --cov-report=xml
 
       - name: Upload coverage report
         uses: codecov/[email protected]

pyproject.toml

@@ -39,6 +39,7 @@ version = { attr = "catlab.__version__" }
 [tool.pytest.ini_options]
 markers = [
     "slow: long-running tests (validation harness determinism), run with `-m slow`",
+    "ui: end-to-end Streamlit AppTest tests (tests/test_app_ui.py)",
 ]
 addopts = "-m \"not slow\""
 

tests/test_app_ui.py

@@ -0,0 +1,273 @@
+# tests/test_app_ui.py
+# End-to-end tests of the Streamlit app (app_ods.py) driven with
+# streamlit.testing.v1.AppTest.
+import io
+import os
+import re
+import sys
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+
+import numpy as np
+import pandas as pd
+import pytest
+from streamlit.testing.v1 import AppTest
+
+from catlab.kinetics_engine import MODEL_NAMES
+
+pytestmark = pytest.mark.ui
+
+APP_PATH = os.path.join(os.path.dirname(__file__), "..", "app_ods.py")
+
+TIME = [0, 15, 30, 45, 60, 90, 120]
+CAT_A = [0, 25, 44, 58, 68, 81, 88]
+CAT_B = [0, 22, 41, 59, 74, 87, 94]
+
+
+def _kinetic_frame():
+    return pd.DataFrame({"Time (min)": TIME, "CatA Removal (%)": CAT_A, "CatB Removal (%)": CAT_B})
+
+
+@pytest.fixture(scope="module")
+def csv_bytes():
+    return _kinetic_frame().to_csv(index=False).encode("utf-8")
+
+
+@pytest.fixture(scope="module")
+def xlsx_bytes():
+    buf = io.BytesIO()
+    with pd.ExcelWriter(buf, engine="openpyxl") as writer:
+        _kinetic_frame().to_excel(writer, sheet_name="Raw_Data", index=False)
+    return buf.getvalue()
+
+
+@pytest.fixture
+def app():
+    return AppTest.from_file(APP_PATH, default_timeout=180).run()
+
+
+def _upload_csv(app, data):
+    app.file_uploader(key="shared_file").upload("data.csv", data, "text/csv").run()
+    return app
+
+
+def _upload_xlsx(app, data):
+    app.file_uploader(key="shared_file").upload(
+        "data.xlsx", data, "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
+    ).run()
+    return app
+
+
+# ================================================================
+# STEP 1 — shared fixtures and smoke tests
+# ================================================================
+
+
+class TestSmoke:
+    def test_app_loads_with_nine_tabs(self, app):
+        assert len(app.tabs) == 9
+        assert not app.exception
+
+    def test_csv_upload_has_no_errors_in_any_tab(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        assert not app.exception
+        assert not [e.value for e in app.error if "Cannot read file" in e.value]
+
+    def test_xlsx_upload_has_no_errors_in_any_tab(self, app, xlsx_bytes):
+        _upload_xlsx(app, xlsx_bytes)
+        assert not app.exception
+        assert not app.error
+
+
+# ================================================================
+# STEP 2 — Tab 1 kinetic fitting
+# ================================================================
+
+
+def _run_analysis(app):
+    [b for b in app.tabs[0].button if "Run" in str(b.label)][0].click().run()
+    return app
+
+
+def _summary(app):
+    for d in app.tabs[0].dataframe:
+        cols = list(d.value.columns)
+        if "Best Model" in cols and "R²" in cols:
+            return d.value
+    return None
+
+
+class TestTabKinetics:
+    def test_summary_has_one_row_per_catalyst(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        _run_analysis(app)
+        summary = _summary(app)
+        assert summary is not None
+        assert list(summary["Catalyst"]) == ["CatA", "CatB"]
+
+    def test_best_model_is_one_of_model_names(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        _run_analysis(app)
+        summary = _summary(app)
+        assert summary is not None
+        for model in summary["Best Model"]:
+            assert model in MODEL_NAMES
+
+    def test_excluding_a_point_drops_points_used_by_one(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        multiselect = app.tabs[0].multiselect(key="excl_CatA Removal (%)")
+        assert "t = 15 min" in multiselect.options
+        multiselect.select("t = 15 min").run()
+        _run_analysis(app)
+        captions = [c.value for c in app.tabs[0].caption]
+        match = re.search(r"CatA: (\d+) points used \(1 excluded\)", "\n".join(captions))
+        assert match is not None
+        # 7 points, minus the t=0 anchor (not counted), minus the excluded point.
+        assert int(match.group(1)) == len(TIME) - 1 - 1
+
+    def test_time_h_column_warns_not_in_minutes(self, app):
+        data = b"Time (h),CatA Removal (%)\n0,0\n1,25\n2,44\n3,58\n4,68\n6,81\n8,88\n"
+        _upload_csv(app, data)
+        assert any("not in minutes" in w.value for w in app.tabs[0].warning)
+
+    def test_t0_row_does_not_change_fit(self, app, csv_bytes):
+        frame = _kinetic_frame()
+        with_t0 = frame.to_csv(index=False).encode("utf-8")
+        without_t0 = frame.iloc[1:].to_csv(index=False).encode("utf-8")
+
+        app_with = AppTest.from_file(APP_PATH, default_timeout=180).run()
+        _upload_csv(app_with, with_t0)
+        _run_analysis(app_with)
+        r2_with = _summary(app_with).set_index("Catalyst")["R²"]
+
+        app_without = AppTest.from_file(APP_PATH, default_timeout=180).run()
+        _upload_csv(app_without, without_t0)
+        _run_analysis(app_without)
+        r2_without = _summary(app_without).set_index("Catalyst")["R²"]
+
+        assert list(r2_with.index) == list(r2_without.index)
+        for cat in r2_with.index:
+            assert np.isclose(r2_with[cat], r2_without[cat], rtol=1e-9)
+
+
+# ================================================================
+# STEP 3 — Tabs 2-9
+# ================================================================
+
+
+class TestTabLinearization:
+    def test_linearization_reports_r2_for_each_catalyst(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        for d in app.tabs[1].dataframe:
+            cols = list(d.value.columns)
+            if set(cols) >= {"Catalyst", "Zero-order", "Pseudo-first"}:
+                pivot = d.value
+                assert list(pivot["Catalyst"]) == ["CatA", "CatB"]
+                for model in ["Zero-order", "Pseudo-first", "Pseudo-second-order", "Elovich"]:
+                    assert model in pivot.columns
+                    assert pivot[model].between(0.0, 1.0).all()
+                return
+        pytest.fail("linearization R² pivot table not found")
+
+
+class TestTabRemoval:
+    def test_removal_renders_efficiency_plots(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        # Streamlit renamed the element type ("imgs" -> "image"); accept both.
+        imgs = [
+            c
+            for c in app.tabs[2].children.values()
+            if getattr(c, "type", None) in ("imgs", "image")
+        ]
+        assert len(imgs) >= 2  # efficiency vs time + final-efficiency bar chart
+        assert not app.exception
+
+
+class TestTabTonTof:
+    def test_option_b_shows_mass_normalized_tof(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        app.tabs[3].radio[0].set_value(
+            "Option B — Carbon-based / Metal-free  (mass-normalized TOF)"
+        ).run()
+        tab = app.tabs[3]
+        cols = {c for d in tab.dataframe for c in d.value.columns}
+        assert "TOF_mass_avg (µmol·g⁻¹·min⁻¹)" in cols
+        assert any("Catalyst_Properties" in i.value for i in tab.info)
+
+
+class TestTabParameterEffect:
+    def test_temperature_sweep_reveals_arrhenius_inputs(self, app):
+        app.tabs[4].selectbox[1].set_value("Temperature (Arrhenius)").run()
+        labels = [n.label for n in app.tabs[4].number_input]
+        assert "k at ref T (min⁻¹)" in labels
+        assert "Eₐ (kJ/mol)" in labels
+
+
+class TestTabOxidant:
+    def test_measured_h2o2_reveals_per_catalyst_inputs(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        app.tabs[5].radio[0].set_value("Yes — I will enter measured consumption").run()
+        tab = app.tabs[5]
+        labels = [n.label for n in tab.number_input]
+        assert any("CatA Removal (%)" in label for label in labels)
+        assert any("CatB Removal (%)" in label for label in labels)
+        cols = {c for d in tab.dataframe for c in d.value.columns}
+        assert "η (%)" in cols
+
+
+class TestTabComparison:
+    def test_comparison_uploads_table_and_offers_axes(self, app):
+        cmp = (
+            b"Experiment,T (C),O/S,kapp (1/min),R2\n"
+            b"Run-1,25,2,0.012,0.989\nRun-2,40,4,0.028,0.994\nRun-3,60,6,0.055,0.997\n"
+        )
+        app.file_uploader(key="cmp_upload").upload("cmp.csv", cmp, "text/csv").run()
+        tab = app.tabs[6]
+        assert tab.dataframe and list(tab.dataframe[0].value["Experiment"]) == [
+            "Run-1",
+            "Run-2",
+            "Run-3",
+        ]
+        labels = [s.label for s in tab.selectbox]
+        assert "X axis" in labels and "Y axis" in labels
+
+
+class TestTabArrhenius:
+    def test_arrhenius_extracts_ea_from_two_temperatures(self, app, csv_bytes):
+        # The 60 °C run must be faster than the 25 °C run; identical files would
+        # give the same k at both temperatures and Ea = 0, testing nothing.
+        fast = pd.DataFrame(
+            {
+                "Time (min)": TIME,
+                "CatA Removal (%)": [0, 45, 68, 81, 89, 96, 98],
+                "CatB Removal (%)": [0, 41, 66, 82, 91, 97, 99],
+            }
+        )
+        fast_bytes = fast.to_csv(index=False).encode("utf-8")
+        app.file_uploader(key="arrhenius_files").set_value(
+            [("t25.csv", csv_bytes, "text/csv"), ("t60.csv", fast_bytes, "text/csv")]
+        ).run()
+        tab = app.tabs[7]
+        tab.number_input(key="arr_T_t25.csv").set_value(25.0)
+        tab.number_input(key="arr_T_t60.csv").set_value(60.0)
+        tab.selectbox(key="arr_model_choice").set_value("Pseudo-first")
+        [b for b in tab.button if "Arrhenius" in str(b.label)][0].click().run()
+        tab = app.tabs[7]
+        assert not tab.error
+        result = tab.dataframe[0].value
+        assert "Eₐ (kJ/mol)" in result.columns
+        assert list(result["n_T"]) == [2, 2]
+        ea = pd.to_numeric(result["Eₐ (kJ/mol)"], errors="coerce")
+        assert (ea > 0).all(), f"faster run at 60 °C must give Ea > 0, got {list(ea)}"
+
+
+class TestTabResiduals:
+    def test_switching_model_changes_r2_metric(self, app, csv_bytes):
+        _upload_csv(app, csv_bytes)
+        tab = app.tabs[8]
+        r2_before = float(tab.metric[0].value)
+        tab.selectbox(key="resid_model").set_value("Pseudo-first").run()
+        r2_after = float(app.tabs[8].metric[0].value)
+        assert r2_before != r2_after
+        assert 0.0 <= r2_before <= 1.0
+        assert 0.0 <= r2_after <= 1.0